Download Report

Title:
Lower limb joint kinetics during the first stance phase in athletics sprinting: three elite athlete
case-studies
Authors:
Neil Edward Bezodisa,b, Aki Ilkka Tapio Saloa, & Grant Trewarthaa
Affiliations and postal addresses:
a
Sport, Health and Exercise Science, University of Bath, UK, BA2 7AY.
b
School of Sport, Health and Applied Science, St Mary’s University College, Twickenham,
UK, TW1 4SX.
Corresponding author: Neil E. Bezodis
email: bezodisn@smuc.ac.uk
Tel: +44 208 240 2325
Co-author: Aki I.T. Salo
email: A.Salo@bath.ac.uk
Tel: +44 1225 38 3569
Co-author: Grant Trewartha
email: G.Trewartha@bath.ac.uk
Tel: +44 1225 38 3055
Abstract:
This study analysed the first stance phase joint kinetics of three elite sprinters to improve the
understanding of technique and investigate how individual differences in technique could
influence the resulting levels of performance. Force (1000 Hz) and video (200 Hz) data were
collected and resultant moments, power and work at the stance leg metatarsal-phalangeal,
ankle, knee and hip joints were calculated. The metatarsal-phalangeal and ankle joints both
exhibited resultant plantarflexor moments throughout stance. Whilst the ankle joint
generated up to four times more energy than it absorbed, the metatarsal-phalangeal joint
was primarily an energy absorber. Knee extensor resultant moments and power were
produced throughout the majority of stance, and the best performing sprinter generated
double and four-times the amount of knee joint energy compared to the other two sprinters.
The hip joint extended throughout stance. Positive hip extensor energy was generated
during early stance before energy was absorbed at the hip as the resultant moment became
flexor dominant towards toe-off. The generation of energy at the ankle appears to be of
greater importance than in later phases of a sprint, whilst knee joint energy generation may
be vital for early acceleration and is potentially facilitated by favourable kinematics at
touchdown.
Keywords: biomechanics, inverse dynamics, kinematics, sprint start, track and field.
Introduction
Athletes must accelerate from a stationary starting position in all athletics sprint events. It
has been demonstrated that exerting as much power as possible from the very start of a
sprint, thus reducing the amount of time spent at sub-maximal velocities early in the race, is
the most favourable strategy for improved overall sprint performance (de Koning, de Groot,
& van Ingen Schenau, 1992; van Ingen Schenau, de Koning, & de Groot, 1994; van Ingen
Schenau, Jacobs, & de Koning, 1991). The start of a sprint is therefore a critically important
element of overall performance, and having left the blocks the first stance phase on the track
is the ground contact where the greatest increase in velocity of any stance phase in a sprint
is achieved (Salo, Keränen, & Viitasalo, 2005). Due to the importance of this first stance
phase, several previous studies have detailed the external kinetic and associated linear
kinematic centre of mass (CM) data from the first stance phase of a sprint (e.g. Mero, 1988;
Mero, Luhtanen, & Komi, 1983; Salo et al., 2005; Slawinski et al., 2010). Results have
shown that higher-performing sprinters produce greater rates of external force development,
resulting in larger net propulsive impulses and thus higher velocities and greater
displacements of the CM than their less well-trained counterparts during the first stance
phase. However, despite the relatively widespread evidence detailing these CM translations,
there is only a limited understanding of the techniques used to achieve them.
To investigate sprint technique in specific detail, stance phase joint kinetics have previously
been reported from groups of sprinters at different distances, from the second step to the
maximum velocity phase (e.g. Bezodis, Kerwin, & Salo, 2008; Brüggemann, Arampatzis,
Emrich, & Potthast, 2008; Burkett, McNamee, & Potthast, 2011; Hunter, Marshall, & McNair,
2004; Jacobs & van Ingen Schenau, 1992; Johnson & Buckley, 2001; Mann, 1981; Mann &
Sprague, 1980). The first stance phase joint kinetics of a single sprinter were recently
described by Charalambous, Irwin, Bezodis and Kerwin (2012) who observed phases of
ankle joint dorsiflexion then plantarflexion and a resultant ankle joint moment which was
plantarflexor dominant throughout stance. The kinetics of the metatarsal-phalangeal (MTP)
joint were not investigated but may be of interest due to previous evidence of MTP
involvement in sprinting (Bezodis, Salo, & Trewartha, 2012; Smith, Lake, Lees, & Worsfold,
2012; Stefanyshyn & Nigg, 1997). Charalambous et al. (2012) also observed hip extension
throughout stance with the resultant joint moment becoming flexor-dominant towards the end
of stance. Whilst these general ankle and hip mechanical patterns are similar to previous
studies from later phases of a sprint, the knee joint results presented by Charalambous et al.
(2012) did not show the flexion during early stance that has been previously observed in
some sprinters at 16 m (Johnson & Buckley, 2001) and all of those studied at 45 m (Bezodis
et al., 2008). However, with the technique of only a single internationally-competitive sprinter
studied during the first stance phase, it remains unclear whether the above joint kinetics
represent a general pattern for elite sprinters. Furthermore, joint kinetics previously reported
from later phases of a sprint show some variation within the groups of sprinters studied (e.g.
Bezodis et al., 2008; Johnson & Buckley, 2001), suggesting that between-sprinter
differences exist. Although these differences in technique may be relatively minor, they are
important to consider in the context of elite sprinting due to the fact that even slight
improvements in performance could give an athlete a competitive edge and affect finishing
position within a race. The aim of this study was therefore two-fold; firstly, to analyse and
understand the common joint kinetics of three elite sprinters during the first stance phase of
a sprint, and secondly, to identify differences in the joint kinetics between these elite
sprinters and therefore gain a better understanding of the differences in their performance
during this important phase of the start.
Methods
Three internationally-competitive sprinters (A: male, 21 years, 82.6 kg, 1.81 m, 100 m
personal best (PB) of 10.14 s; B: 20 years, 86.9 kg, 1.78 m, 100 m PB of 10.28 s; C: female,
26 years, 60.5 kg, 1.76 m, 100 m hurdle PB of 12.72 s) provided written informed consent to
participate in this study which had received local research ethics committee approval. All of
the sprinters had reached the final of European or World Indoor Championships less than
two years prior to this study. The cohort contained sprinters of both sexes to provide wider
potential applications as performance can be considered as a physical characteristic rather
than a sex-dependent issue. Each sprinter completed a series of five (A and B) or four (C)
maximal effort 30 m sprints from their chosen starting block settings on an indoor track.
These sprints were completed as a normal part of their training just prior to the competition
phase of the indoor season. All sprints were initiated by an experienced starter who provided
standard starting commands before pressing a trigger button which provided an auditory
stimulus for the sprinters to start. After each trial, sprinters were allowed their normal
recovery (about 8-10 minutes) in order to facilitate performance without the effects of fatigue.
The blocks were located so that the first stance phase after block exit would occur near the
centre of a force platform (9287BA, Kistler, Switzerland) operating at 1000 Hz and covered
with a standard artificial track surface to be flush with the remainder of the track. A highspeed video camera (Motion-Pro HS-1, Redlake, USA) operating at 200 Hz was located
perpendicular to the direction of the running lane, 0.95 m in front of the start line and 25 m
from the lane centre. An area of 2.00 m horizontally × 1.60 m vertically was calibrated using
a rectangular calibration frame positioned inside a field of view 2.50 m wide. The camera
collected images at a resolution of 1280 × 1024 pixels with a 1/1000 s shutter speed. A
second high-speed video camera (Motion-Pro HS-1, Redlake, USA) was set exactly as
above aside from being 0.25 m ahead the start line. This camera was only used for
determining CM velocity at the onset of the first stance phase. The video and force data
were synchronised to the nearest millisecond with the aforementioned trigger button which
sent signals to the force platform and a series of 20 LEDs illuminating at 1 ms intervals in the
view of the camera.
Touchdown and toe-off were identified from the raw force data using a threshold of 2
standard deviations above the zero load force platform data. Twenty specific anatomical
points (vertex and C7, bilateral shoulder, elbow, wrist, hip, knee, ankle and second MTP joint
centres, fingertips and distal toes) were then manually digitised (Peak Motus® v. 8.5, Vicon,
UK) from 10 frames prior to touchdown until 10 frames after toe-off. All subsequent data
analysis was undertaken using custom-developed scripts in Matlab™ (v. 7.4.0, The
MathWorks™, USA). Where the hand left the field of view during the last few of the 10
frames after toe-off in some of the trials, these trajectories were padded via reflection prior to
filtering in order to reduce the effects of any endpoint error (Smith, 1989). The raw coordinates were scaled (using projective scaling based on the four corner points of the
calibration frame) and the resulting joint centre displacement data were digitally filtered using
a fourth-order Butterworth digital filter with a cut-off frequency of 24 Hz (selected as the
mean value from residual analyses of all stance leg joint centre trajectories from all trials).
These filtered displacement data were then combined with individual specific segmental
inertia data calculated from 95 direct anthropometric measurements (Yeadon, 1990) taken
on each sprinter by an experienced researcher to create a 16 segment model (head, trunk,
upper arms, lower arms, hands, thighs, shanks, rearfeet, forefeet). To account for the mass
of the spiked shoes, 0.20 kg was added to the mass of each foot (e.g. Hunter et al., 2004).
The division of spike mass between the two foot segments was determined directly from the
ratio of forefoot:rearfoot length, assuming an equal division of mass across the length of the
spike. The whole body CM displacement time-history was consequently calculated from the
segmental data using a summation of segmental moments approach. Where required, the
linear and angular displacement time-histories were subjected to second central difference
calculations (Miller & Nelson, 1973) in order to derive their corresponding velocity and
acceleration time-histories. Extension at the hip and knee joints, and plantarflexion at the
ankle and MTP joints, were defined as positive.
Two approaches to filtering the force data were adopted (Bezodis, Salo, & Trewartha, 2013).
A cut-off frequency of 150 Hz was used when determining discrete force values and average
horizontal external power. This power value was used as the measure of first stance phase
performance due to the aforementioned importance of the production of maximal external
power from the very start of a sprint for optimum overall sprint performance (de Koning et al.
1992; van Ingen Schenau et al. 1991; 1994). Average horizontal external power incorporates
both the change in velocity and the time taken for these changes into a single objective
measure and is quantified based on the rate of change in kinetic energy of the CM in a
horizontal direction (Bezodis, Salo, & Trewartha, 2010). The initial velocity used in this
calculation was determined from the second high-speed camera using the procedures
described by Bezodis et al. (2010). Quantifying average horizontal external power during just
the first stance phase therefore provides a performance measure that is directly relevant to
the analysed technique. For the kinetic inputs to the inverse dynamics analysis, the force
platform data were downsampled to the sampling rate of the kinematic data (200 Hz), and
filtered with a cut-off frequency of 24 Hz to prevent the generation of artefacts soon after
impact (Bezodis et al., 2013; Bisseling & Hof, 2006; Kristianslund, Krosshaug, & van den
Bogert, 2012). Using these filtered kinematic and kinetic data in combination with the
individual-specific segmental inertia data, a 2D inverse dynamics analysis was undertaken
(Elftman, 1939; Winter, 2005). Resultant joint moments were calculated about the stance leg
MTP, ankle, knee and hip joints. Joint powers were then calculated as the product of the
resultant joint moment and joint angular velocity, and phases of positive and negative power
at each joint were identified. To quantify energy absorption and generation, joint work was
calculated as the time-integral (trapezium rule) of each major positive and negative phase of
the joint power time-histories (see Figures 1 to 4 for illustrations of these power phases). To
facilitate comparisons between the sprinters, data were normalised to dimensionless
numbers according to the convention of Hof (1996), with the power normalisation adjusted
as outlined by Bezodis et al. (2010). Specifically, angular velocity data were divided by
(gravity / leg length)1/2, force data were divided by weight, moment and work data were
divided by (weight × leg length), and power data were divided by (body mass × gravity3/2 ×
leg length1/2). Individual mean time-histories were presented to address the first part of the
aim and understand common joint kinetic patterns. To address the second aim of this study
and quantitatively identify any between-sprinter differences, confidence intervals (95%) for
discrete joint kinetic variables were calculated using the appropriate critical values of t at the
two-tailed level for each of the three sprinters (Thomas & Nelson, 2001). Due to these
confidence intervals representing the likely range of the true value, between-sprinter
differences were identified where confidence limits did not overlap.
Results
Sprinter B exhibited the highest normalised average horizontal external power (a measure of
first stance phase performance) of 1.030 ± 0.038 (mean ± 95% confidence interval; sprinter
A = 0.833 ± 0.070, sprinter C = 0.790 ± 0.035; Table 1). All three sprinters exhibited a
braking phase of no more than 0.017 s during which sprinter B exhibited the lowest braking
forces. This was followed by a propulsive phase which always lasted for at least 90% of
stance, and during which sprinter B exhibited the greatest propulsive forces (Table 1).
****Table 1 near here****
The joint angle, angular velocity, resultant moments and powers at each of the four leg joints
during stance are presented in Figures 1 to 4. Both the MTP and ankle joints exhibited
resultant plantarflexor moments throughout stance (Figures 1c and 2c). The ankle showed a
clear pattern of dorsiflexion then plantarflexion (Figures 2a and 2b), and thus phases of
energy absorption (A1) then generation (A2; Figure 2d) with sprinter B generating more
energy than sprinters A and C during phase A2 (Table 1). At the MTP joint, there were
fluctuations in joint power during the first half of stance, then a phase of energy absorption
as the joint dorsiflexed from around mid-stance (M1) before a phase of energy generation as
the joint plantarflexed towards toe-off (M2; Figure 1d). The knee and hip joints extended
from touchdown throughout the majority of stance (Figures 3a and 4a) and in several trials
began to flex just before toe-off. Sprinters A and C exhibited a consistent knee flexor
resultant moment at touchdown and thus an initial phase of energy absorption (K1). Knee
joint moments became extensor dominant and energy was thus generated (K2) until late
stance where energy was absorbed (K3; Figures 3c and 3d). Sprinter B exhibited resultant
knee extensor moments and energy generation from touchdown and generated 363% and
188% of the normalised energy at the knee joint that sprinters A and C produced during
phase K2, respectively (Table 1). At the hip joint, large resultant extensor moments were
evident at touchdown before these gradually reduced and became flexor dominant later in
stance (Figure 4c). The hip joint therefore generated energy during early stance (H1) and
absorbed it during late stance (H2; Figure 4d).
****Figures 1-4 near here****
Discussion
The results of this study demonstrated a common joint kinetic pattern during the first stance
phase of a maximal effort sprint in three elite sprinters. The similarities in joint kinetics
between these three sprinters and with the single internationally-competitive sprinter
previously analysed by Charalambous et al. (2012) yield confidence in the application of
these data to understanding elite athletes' first stance phase performance. Additionally,
some differences were identified between the three sprinters in the current study, and these
can be explored to understand how they may relate to the performance differences in this
first stance phase.
The existence of a resultant plantarflexor MTP moment throughout stance (Figure 1c)
concurs with previously recorded MTP joint moments in sprinting (Smith et al., 2012;
Stefanyshyn & Nigg, 1997). These moments, and in particular the energy absorbed by the
MTP joint during mid-late stance (phase M1), reiterate the importance of considering the
MTP joint in kinetic analyses of sprinting as highlighted by Bezodis et al. (2012). When
comparing sprinters, Sprinter B generated a higher normalised MTP plantarflexor moment
than sprinters A and C (Table 1). This may be an important factor in his higher level of sprint
performance as it has been suggested by Goldmann and Brüggemann (2012) and
Goldmann, Sanno, Willwacher, Heinrich and Brüggemann (2013) that toe flexor muscles are
important contributors to movements where an individual is in a forward leaning position (as
is the case during early acceleration) and strength training of these muscles can improve
performance in tasks which require a forward lean (i.e. horizontal jumping; Goldmann et al.,
2013). It is also possible that a greater contribution from the biarticular toe flexor muscles
(flexor digitorum longus and flexor hallucis longus) could consequently have allowed Sprinter
B to generate greater positive plantarflexor energy at the ankle joint than sprinters A and C
(Table 1). These muscles also help to reduce ankle dorsiflexion during running (Mann &
Hagy, 1979), and previous research has shown that stiffening the shoe around the MTP joint
can significantly increase the resultant plantarflexor ankle joint moment (Roy & Stefanyshyn,
2006). It must be acknowledged that the specific components of the resultant plantarflexor
MTP joint moments cannot be identified using the current inverse dynamics analysis. Whilst
greater resultant plantarflexor MTP joint moments could be due to the toe flexor muscles,
they may also be due to passive biological structures (e.g. plantar fascia) or a non-biological
component such as the stiffness of the spiked shoes (Stefanyshyn & Fusco, 2004; Toon,
Vinet, Pain, & Caine, 2011; Willwacher, König, Potthast, & Brüggemann, in press).
Whilst the ankle joint kinetics were broadly similar to other phases of a sprint in terms of a
resultant plantarflexor moment being evident throughout stance and phases of energy
absorption followed by generation (Figures 2c and 2d), closer consideration of the joint work
data highlights some important differences. Energy was generated at the ankle from around
mid-stance onwards (phase A2), and sprinters A, B and C generated more energy at the
ankle than they absorbed (phase A1) by a factor of 3.4, 3.3 and 4.0, respectively (Table 1
and Figure 2d). Johnson and Buckley (2001) previously presented ankle joint power timehistories from the 16 m mark which showed energy absorption of similar magnitude to the
subsequent energy generation (i.e. a factor of 1.0), whilst at maximum velocity Bezodis et al.
(2008) found the amount of energy absorbed exceeded that generated (whole group mean
factor = 0.6). It therefore appears that the net energy generated at the ankle joint gradually
decreases from a large positive value to a negative value as a sprint progresses. Combined
with the fact that sprinter B generated more ankle joint energy in phase A2 than sprinters A
and C, an ability to generate ankle joint energy therefore appears to be an important aspect
of early acceleration technique.
In comparison to the knee joint at maximum velocity (Bezodis et al., 2008) large magnitudes
of extensor resultant moment and positive power were evident at the knee joint throughout
the majority of stance (Figures 3c and 3d). This highlights the relative importance of the knee
joint during early acceleration, similar to Charalambous et al. (2012). In the maximum
velocity phase, the knee extensors are initially used to terminate the negative vertical
velocity of the CM at touchdown before playing a role in generating positive vertical and
horizontal velocity for the remainder of stance (Mann, 1981). As there is a lower initial
negative vertical velocity to be reversed during the start of a sprint, the knee is therefore able
to have an increased role in the generation of positive power, and thus acceleration of a
sprinter. Consequently, it is of interest that some of the most evident between-sprinter
differences were observed at the knee. The greater energy generated at the knee joint by
sprinter B (363% and 188% of that generated by sprinters A and C during phase K2,
respectively; Table 1) was a result of both the earlier rise and the higher peak in the resultant
knee joint moment (joint extension velocities were more similar between sprinters than
resultant moments; Figure 3b). In accordance with the geometrical constraints identified by
van Ingen Schenau, Bobbert and Rozendal (1987), the knee joint has considerable potential
to contribute to forwards horizontal translation of the CM at the start of a sprint, and is thus a
likely factor in the generation of higher levels of external power by sprinter B. The ability of
sprinter B to generate knee extensor resultant moments from the onset of stance may also
have been assisted by his touchdown kinematics, as he exhibited a lower horizontal toe
velocity at touchdown compared to sprinters A and C (Table 1). This has previously been
associated with reduced braking force magnitude (Jacobs & van Ingen Schenau, 1992;
Mann & Herman, 1985; Putnam & Kozey, 1989) and thus the touchdown kinematics of
sprinter B helped to reduce the magnitude of the braking forces he experienced (Table 1)
and assisted the generation of energy at the knee joint during stance. Experimental training
interventions or computer simulations of the first stance phase may be useful to investigate
the extent to which manipulating touchdown kinematics can influence knee joint kinetics and
ultimately performance during early acceleration.
The resultant hip joint extensor moments observed at touchdown for all athletes in this study,
and which peaked soon after touchdown (Figure 4c), are consistent with the later phases of
a sprint (Bezodis et al., 2008; Johnson & Buckley, 2001). These large hip extensor resultant
moments at touchdown imply that an extensor resultant moment was present at the hip prior
to touchdown. This suggests the existence of a similar strategy to that used in later phases
of a sprint, whereby the hip extensor moment reduces the forward momentum of the swing
leg prior to touchdown, decreasing the toe touchdown velocity and thus the braking forces
experienced (Jacobs & van Ingen Schenau, 1992; Mann & Herman, 1985; Putnam & Kozey,
1989). The resultant hip moment became flexor dominant before toe-off to absorb energy,
thus reducing hip extension and helping to terminate ground contact. Sprinter B performed
greater negative work at the hip joint during this phase (H2) compared to sprinters A and C
which suggests that this strategy which has previously been highlighted as important in
maximum velocity sprinting (Mann, 1981; Mann & Sprague, 1980) could also play an
important role in early acceleration.
It was decided to limit the current analysis to sprinters who had competed in a major
international final within the last two years, and this study thus analysed the techniques of
three world-class sprinters. Whilst a larger cohort of sprinters would be desirable, there exist
low numbers of potential elite participants due to the inherent nature of being elite.
Combining sub-elite sprinters in to the current study could have weakened the
understanding that could be obtained. Data were collected outside of competition to enable
the collection of ground reaction force data, but were collected as close to the competition
season as possible when the performance levels of the sprinters were intended to peak. This
limited the number of trials and measurements that could be collected, and whilst additional
data such as joint moment-velocity relationships or muscle architecture and tendon moment
arms (e.g. Baxter, Novack, van Werkhoven, Pennell, & Piazza, 2012; Karamanidis et al.,
2011; Lee & Piazza, 2009) could further the understanding of some of the inter-individual
differences observed, the current data clearly show some consistent joint kinetic patterns
and thus offer a valuable insight in to the techniques exhibited by sprinters who were
competing at the highest level at the time of this study.
Conclusion
This study identified a clear joint kinetic pattern associated with the four main stance leg
joints during the first stance phase of a sprint. The MTP joint was a net absorber of energy,
whereas at the ankle joint net energy generation was evident and considerably greater in
comparison with previously published data from later phases of a sprint. The hip joint was
more similar to other phases of a sprint as it extended throughout stance. It initially
generated energy before energy was absorbed as the resultant joint moment became flexor
after mid-stance, similar to the maximum velocity phase of sprinting. Finally, the knee joint
was a clear net energy generator. The ability to generate a large resultant knee extensor
moment appeared to be a key difference between the highest performing sprinter and the
other sprinters in this study. An earlier rise and a greater peak in this resultant knee extensor
moment may have been facilitated by the kinematics of this sprinter at touchdown, due to a
reduced forwards velocity of the foot which led to lower braking forces. Enhancing knee
extensor power therefore has the potential to improve early acceleration performance, either
through technical adjustments around touchdown or through strengthening the relevant
musculature.
Acknowledgements
The authors are grateful to the coaches and athletes for allowing data to be collected at their
training sessions, and to Dr Ian Bezodis and the Cardiff School of Sport, UK, for their
assistance with data collection and facilities. The University of Bath, UK, and UK Athletics
provided part funding for the study.
References
Baxter, J. R., Novack, T. A., van Werkhoven, H., Pennell, D. R., & Piazza, S. J. (2012).
Ankle joint mechanics and foot proportions differ between human sprinters and nonsprinters. Proceedings of the Royal Society B, 279, 2018-2024.
Bezodis, I. N., Kerwin, D. G., & Salo, A. I. T. (2008). Lower-limb mechanics during the
support phase of maximum-velocity sprint running. Medicine and Science in Sports
and Exercise, 40, 707-715.
Bezodis, N. E., Salo, A. I. T., & Trewartha, G. (2010). Choice of sprint start performance
measure affects the performance-based ranking within a group of sprinters: which is
the most appropriate measure? Sports Biomechanics, 9, 258-269.
Bezodis, N. E., Salo, A. I. T., & Trewartha, G. (2012). Modelling the stance leg in 2D
analyses of sprinting: inclusion of the MTP joint affects joint kinetics. Journal of
Applied Biomechanics, 28, 222-227.
Bezodis, N. E., Salo, A. I. T., & Trewartha, G. (2013). Excessive fluctuations in knee joint
moments during early stance in sprinting are caused by digital filtering procedures.
Gait & Posture, 38, 653-657.
Bisseling, R. W., & Hof, A. L. (2006). Handling of impact forces in inverse dynamics. Journal
of Biomechanics, 39, 2438-2444.
Brüggemann, G. -P., Arampatzis, A., Emrich, F., & Potthast, W. (2009). Biomechanics of
double transtibial amputee sprinting using dedicated sprinting prostheses. Sports
Technology, 1, 220-227.
Burkett, B., McNamee, M., & Potthast, W. (2011). Shifting boundaries in sports technology
and disability: equal rights or unfair advantage in the case of Oscar Pistorius?
Disability & Society, 26, 643-654.
Charalambous, L., Irwin, G., Bezodis, I. N., & Kerwin, D. (2012). Lower limb joint kinetics and
ankle joint stiffness in the sprint start push-off. Journal of Sports Sciences, 30, 1-9.
de Koning, J. J., de Groot, G., & van Ingen Schenau, G. J. (1992). A power equation for the
sprint in speed skating. Journal of Biomechanics, 25, 573-580.
Elftman, H. (1939). Forces and energy changes in the leg during walking. American Journal
of Physiology, 125, 339-356.
Goldmann, J. -P., & Brüggemann, G. -P. (2012). The potential of human toe flexor muscles
to produce force. Journal of Anatomy, 221, 187-194.
Goldmann, J. -P., Sanno, M., Willwacher, S., Heinrich, K., & Brüggemann, G. -P. (2013).
The potential of toe flexor muscles to enhance performance. Journal of Sports
Sciences, 31, 424-433.
Hof, A. L. (1996). Scaling gait data to body size. Gait & Posture, 4, 222-223.
Hunter, J. P., Marshall, R. N., & McNair, P. J. (2004). Segment-interaction analysis of the
stance limb in sprint running. Journal of Biomechanics, 37, 1439-1446.
Jacobs, R., & van Ingen Schenau, G. J. (1992). Intermuscular coordination in a sprint pushoff. Journal of Biomechanics, 25, 953-965.
Johnson, M. D., & Buckley, J. G. (2001). Muscle power patterns in the mid-acceleration
phase of sprinting. Journal of Sports Sciences, 19, 263-272.
Karamanidis, K., Albracht, K., Braunstein, B., Catala, M. M., Goldmann, J. -P., &
Brüggemann, G. -P. (2011). Lower leg musculoskeletal geometry and sprint
performance. Gait & Posture, 34, 138-141.
Kristianslund, E., Krosshaug, T., & van den Bogert, A. J. (2012). Effect of low pass filtering
on joint moments from inverse dynamics: implications for injury prevention. Journal
of Biomechanics, 45, 666-671.
Lee, S. S. M., & Piazza, S. J. (2009). Built for speed: musculoskeletal structure and sprinting
ability. Journal of Experimental Biology, 212, 3700-3707.
Mann, R. V. (1981). A kinetic analysis of sprinting. Medicine and Science in Sports and
Exercise, 13, 325-328.
Mann, R., & Herman, J. (1985). Kinematic analysis of Olympic sprint performance: men's
200 meters. International Journal of Sports Biomechanics, 1, 151-162.
Mann, R., & Sprague, P. (1980). A kinetic analysis of the ground leg during sprint running.
Research Quarterly for Exercise and Sport, 51, 334-348.
Mann, R. A., & Hagy, J. L. (1979). The function of the toes in walking, jogging and running.
Clinical Orthopaedics and Related Research, 142, 24-29.
Mero, A. (1988). Force-time characteristics and running velocity of male sprinters during the
acceleration phase of sprinting. Research Quarterly for Exercise and Sport, 59, 9498.
Mero, A., Luhtanen, P., & Komi, P. V. (1983). A biomechanical study of the sprint start.
Scandinavian Journal of Sports Science, 5, 20-28.
Miller, D., & Nelson, R. (1973). Biomechanics of sport: A research approach. Philadelphia,
PA: Lea and Febiger.
Putnam, C. A., & Kozey, J. W. (1989). Substantive issues in running. In C. L. Vaughan (Ed.),
Biomechanics of Sport (pp. 1-33). Boca Raton, FL: CRC Press.
Roy, J. P. R., & Stefanyshyn, D. J. (2006). Shoe midsole longitudinal bending stiffness and
running economy, joint energy, and EMG. Medicine and Science in Sports and
Exercise, 38, 562-569.
Salo, A. I. T., Keränen, T., & Viitasalo, J. T. (2005). Force production in the first four steps of
sprint running. In Q. Wang (Ed.), Proceedings of XXIII International Symposium on
Biomechanics in Sports (pp. 313-317). Beijing, China: The China Institute of Sport
Science.
Slawinski, J., Bonnefoy, A., Levêque J.-M., Ontanon, G., Riquet, A., Dumas, R., & Chèze, L.
(2010). Kinematic and kinetic comparisons of elite and well-trained sprinters during
sprint start. Journal of Strength and Conditioning Research, 24, 896-905.
Smith, G. (1989). Padding point extrapolation techniques for the Butterworth digital filter.
Journal of Biomechanics, 22, 967-971.
Smith, G., Lake, M., Lees, A., & Worsfold, P. (2012). Measurement procedures affect the
interpretation of metatarsophalangeal joint function in accelerated sprinting. Journal
of Sports Sciences, 30, 1521-1527.
Stefanyshyn, D., & Fusco, C. (2004). Increased shoe bending stiffness increases sprint
performance. Sports Biomechanics, 3, 55-66.
Stefanyshyn, D. J., & Nigg, B. M. (1997). Mechanical energy contribution of the
metatarsophalangeal joint to running and sprinting. Journal of Biomechanics, 30,
1081-1085.
Thomas, J. R., & Nelson, J. K. (2001). Research methods in physical activity (4th ed.).
Champaign, IL: Human Kinetics.
Toon, D., Vinet, A., Pain, M. T. G., & Caine, M. P. (2011). A methodology to investigate the
relationship between lower-limb dynamics and shoe stiffness using custom-built
footwear. Journal of Sports Engineering and Technology, 225, 32-37.
van Ingen Schenau, G. J., Bobbert, M. F., & Rozendal, R. H. (1987). The unique action of biarticular muscles in complex movements. Journal of Anatomy, 155, 1-5.
van Ingen Schenau, G. J., de Koning, J. J., & de Groot, G. (1994). Optimisation of sprinting
performance in running, cycling and speed skating. Sports Medicine, 17, 259-275.
van Ingen Schenau, G. J., Jacobs, R., & de Koning J. J. (1991). Can cycle power predict
sprint running performance. European Journal of Applied Physiology and
Occupational Physiology, 63, 255-260.
Willwacher, S., Konig, M., Potthast, W., & Brüggemann, G.–P. (in press). Does specific
footwear facilitate energy storage and return at the metatarso phalangeal joint in
running? Journal of Applied Biomechanics.
Winter, D. A. (2005) Biomechanics and motor control of human movement (3rd ed.). New
York, NY: Wiley.
Yeadon, M. R. (1990). The simulation of aerial movement-II. A mathematical inertia model of
the human-body. Journal of Biomechanics, 23, 67-74.
Table 1. Ground reaction force and joint kinetic discrete variables for each of the three sprinters (mean and 95% confidence limits).
A
B
0.833 (0.764 – 0.903)
B
Peak braking force
0.903 (0.625 – 1.180)
B
Duration of braking phase (s)
0.016 (0.016 – 0.017)
Average horizontal external power
B
Peak propulsive force
1.158 (1.078 – 1.237)
Duration of propulsive phase (s)
0.176 (0.166 – 0.187)
B
Vertical impact peak
1.494 (1.356 – 1.633)
Vertical active peak
2.094 (1.938 – 2.250)
Peak MTP plantar flexor moment
0.138 (0.130 – 0.146)
Peak ankle plantar flexor moment
0.452 (0.421 – 0.483)
B
C
B
C
1.030 (0.992 – 1.068)
A,C
0.790 (0.755 – 0.825)
B
0.172 (0.056 – 0.288)
A,C
0.525 (0.346 – 0.705)
B
0.016 (0.015 – 0.017)
1.284 (1.260 – 1.307)
1.064 (1.035 – 1.092)
B
C
0.161 (0.152 – 0.170)
B
A,C
1.501 (1.402 – 1.599)
B
C
2.021 (1.917 – 2.126)
B
A,C
0.120 (0.103 – 0.138)
B
0.175 (0.172 – 0.179)
0.709 (0.599 – 0.820)
2.178 (2.154 – 2.202)
0.177 (0.165 – 0.190)
0.451 (0.443 – 0.460)
C
0.378 (0.353 – 0.404)
A,B
A,C
0.138 (0.098 – 0.178)
B
Peak knee extensor moment
0.087 (0.053 – 0.121)
Peak hip extensor moment
0.359 (0.279 – 0.439)
0.354 (0.295 – 0.414)
C
0.273 (0.256 – 0.291)
B
Peak hip flexor moment
0.390 (0.323 – 0.457)
0.432 (0.347 – 0.518)
C
0.297 (0.268 – 0.326)
B
Peak negative MTP power
0.307 (0.283 – 0.330)
A,C
0.255 (0.185 – 0.324)
B
Peak positive MTP power
0.261 (0.035 – 0.487)
B
Peak negative ankle power
0.419 (0.289 – 0.550)
Peak positive ankle power
1.206 (1.109 – 1.302)
Peak negative knee power (early stance)
0.362 (0.282 – 0.443)
B
B,C
B
0.216 (0.187 – 0.245)
0.015 (0.014 – 0.017)
A,C
0.442 (0.393 – 0.492)
0.188 (0.059 – 0.177)
0.146 (0.046 – 0.245)
0.418 (0.309 – 0.528)
0.363 (0.180 – 0.545)
1.488 (1.454 – 1.523)
A
0.000 (0.000 – 0.000)
0.422 (0.374 – 0.470)
A,C
0.249 (0.224 – 0.274)
B
C
0.258 (0.186 – 0.331)
B
Peak positive knee power
0.148 (0.058 – 0.237)
Peak negative knee power (late stance)
0.206 (0.075 – 0.336)
0.127 (0.075 – 0.179)
Peak positive hip power
1.450 (1.097 – 1.804)
1.185 (0.929 – 1.442)
B
Peak negative hip power
0.535 (0.268 – 0.803)
Total negative MTP work (M1)
0.042 (0.026 – 0.058)
B
Total positive MTP work (M2)
0.008 (0.005 – 0.011)
Total negative ankle work (A1)
0.052 (0.037 – 0.068)
Total positive ankle work (A2)
Total negative knee work in early stance (K1)
Total positive knee work (K2)
0.175 (0.156 – 0.195)
B
0.025 (0.014 – 0.036)
B,C
0.030 (0.022 – 0.039)
B,C
1.251 (1.002 – 1.500)
A,C
0.870 (0.804 – 0.936)
A,C
0.061 (0.054 – 0.067)
0.003 (0.002 – 0.003)
A
0.067 (0.056 – 0.079)
0.223 (0.213 – 0.232)
A,B
0.842 (0.568 – 1.115)
0.474 (0.258 – 0.691)
B
0.043 (0.031 – 0.056)
0.004 (0.000 – 0.007)
0.041 (0.022 – 0.061)
A,C
0.163 (0.138 – 0.188)
B
C
0.001 (0.000 – 0.002)
A
0.000 (0.000 – 0.000)
0.109 (0.087 – 0.130)
0.087 (0.054 – 0.120)
A,C
0.058 (0.039 – 0.076)
A,B
Total negative knee work in late stance (K3)
0.020 (0.003 – 0.037)
Total positive hip work (H1)
0.191 (0.140 – 0.242)
Total negative hip work (H2)
0.038 (0.017 – 0.058)
Touchdown distance
0.090 (0.053 – 0.126)
Touchdown velocity
2.293 (1.506 – 3.079)
B
B,C
0.010 (0.003 – 0.017)
0.215 (0.187 – 0.243)
0.068 (0.062 – 0.075)
A,C
0.048 (0.019 – 0.077)
0.003 (-0.178 – 0.184)
A
0.025 (0.008 – 0.041)
0.180 (0.138 – 0.222)
0.033 (0.022 – 0.044)
B
0.055 (0.031 – 0.079)
0.779 (0.132 – 1.425)
A
All normalised values are dimensionless – see methods section for details of normalisation procedures. Superscript letters represent non-overlapping
confidence intervals with the respective sprinter indicated by the letter. Abbreviations in parentheses with the joint work data correspond to the phases of
positive and negative work illustrated in Figures 1 to 4.
Figure 1. Mean MTP a) joint angle, b) normalised angular velocity, c) normalised resultant
joint moment and d) normalised joint power time-histories for sprinters A (dotted line), B
(dashed line) and C (solid line).
Figure 2. Mean ankle a) joint angle, b) normalised angular velocity, c) normalised resultant
joint moment and d) normalised joint power time-histories for sprinters A (dotted line), B
(dashed line) and C (solid line).
Figure 3. Mean knee a) joint angle, b) normalised angular velocity, c) normalised resultant
joint moment and d) normalised joint power time-histories for sprinters A (dotted line), B
(dashed line) and C (solid line).
Figure 4. Mean hip a) joint angle, b) normalised angular velocity, c) normalised resultant joint
moment and d) normalised joint power time-histories for sprinters A (dotted line), B (dashed
line) and C (solid line).